Geometry of goodness-of-fit testing in high dimensional low sample size modelling

Paul Marriott; Radka Sabolova; Germain van Bever; Frank Critchley

doi:10.1007/978-3-319-25040-3_61

Geometry of goodness-of-fit testing in high dimensional low sample size modelling

Paul Marriott, Radka Sabolova, Germain van Bever, Frank Critchley

Résultats de recherche: Contribution dans un livre/un catalogue/un rapport/dans les actes d'une conférence › Article dans les actes d'une conférence/un colloque

Résumé

We introduce a new approach to goodness-of-fit testing in the high dimensional, sparse extended multinomial context. The paper takes a computational information geometric approach, extending classical higher order asymptotic theory. We show why the Wald – equivalently, the Pearson χ2 and score statistics – are unworkable in this context, but that the deviance has a simple, accurate and tractable sampling distribution even for moderate sample sizes. Issues of uniformity of asymptotic approximations across model space are discussed. A variety of important applications and extensions are noted.

langue originale	Anglais
titre	Geometric Science of Information - 2nd International Conference, GSI 2015, Proceedings
rédacteurs en chef	Frank Nielsen, Frank Nielsen, Frank Nielsen, Frederic Barbaresco, Frederic Barbaresco, Frank Nielsen
Editeur	Springer Verlag
Pages	569-576
Nombre de pages	8
ISBN (imprimé)	9783319250397, 9783319250397
Les DOIs	https://doi.org/10.1007/978-3-319-25040-3_61
Etat de la publication	Publié - 2015
Modification externe	Oui
Evénement	2nd International Conference on Geometric Science of Information, GSI 2015 - Palaiseau, France Durée: 28 oct. 2015 → 30 oct. 2015

Série de publications

Nom	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	9389
ISSN (imprimé)	0302-9743
ISSN (Electronique)	1611-3349

Une conférence

Une conférence	2nd International Conference on Geometric Science of Information, GSI 2015
Pays/Territoire	France
La ville	Palaiseau
période	28/10/15 → 30/10/15

Accès au document

10.1007/978-3-319-25040-3_61

Autres fichiers et liens

Link to publication in Scopus

Contient cette citation

Marriott, P., Sabolova, R., van Bever, G., & Critchley, F. (2015). Geometry of goodness-of-fit testing in high dimensional low sample size modelling. Dans F. Nielsen, F. Nielsen, F. Nielsen, F. Barbaresco, F. Barbaresco, & F. Nielsen (eds.), Geometric Science of Information - 2nd International Conference, GSI 2015, Proceedings (p. 569-576). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol 9389). Springer Verlag. https://doi.org/10.1007/978-3-319-25040-3_61

Marriott, Paul ; Sabolova, Radka ; van Bever, Germain et al. / Geometry of goodness-of-fit testing in high dimensional low sample size modelling. Geometric Science of Information - 2nd International Conference, GSI 2015, Proceedings. Editeur / Frank Nielsen ; Frank Nielsen ; Frank Nielsen ; Frederic Barbaresco ; Frederic Barbaresco ; Frank Nielsen. Springer Verlag, 2015. p. 569-576 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{7af02731e62d412ea33ef1dc38a7671f,

title = "Geometry of goodness-of-fit testing in high dimensional low sample size modelling",

abstract = "We introduce a new approach to goodness-of-fit testing in the high dimensional, sparse extended multinomial context. The paper takes a computational information geometric approach, extending classical higher order asymptotic theory. We show why the Wald – equivalently, the Pearson χ2 and score statistics – are unworkable in this context, but that the deviance has a simple, accurate and tractable sampling distribution even for moderate sample sizes. Issues of uniformity of asymptotic approximations across model space are discussed. A variety of important applications and extensions are noted.",

author = "Paul Marriott and Radka Sabolova and {van Bever}, Germain and Frank Critchley",

year = "2015",

doi = "10.1007/978-3-319-25040-3_61",

language = "English",

isbn = "9783319250397",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "569--576",

editor = "Frank Nielsen and Frank Nielsen and Frank Nielsen and Frederic Barbaresco and Frederic Barbaresco and Frank Nielsen",

booktitle = "Geometric Science of Information - 2nd International Conference, GSI 2015, Proceedings",

address = "Germany",

note = "2nd International Conference on Geometric Science of Information, GSI 2015 ; Conference date: 28-10-2015 Through 30-10-2015",

}

Marriott, P, Sabolova, R, van Bever, G & Critchley, F 2015, Geometry of goodness-of-fit testing in high dimensional low sample size modelling. Dans F Nielsen, F Nielsen, F Nielsen, F Barbaresco, F Barbaresco & F Nielsen (eds), Geometric Science of Information - 2nd International Conference, GSI 2015, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), VOL. 9389, Springer Verlag, p. 569-576, 2nd International Conference on Geometric Science of Information, GSI 2015, Palaiseau, France, 28/10/15. https://doi.org/10.1007/978-3-319-25040-3_61

Geometry of goodness-of-fit testing in high dimensional low sample size modelling. / Marriott, Paul; Sabolova, Radka; van Bever, Germain et al.
Geometric Science of Information - 2nd International Conference, GSI 2015, Proceedings. Ed. / Frank Nielsen; Frank Nielsen; Frank Nielsen; Frederic Barbaresco; Frederic Barbaresco; Frank Nielsen. Springer Verlag, 2015. p. 569-576 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol 9389).

Résultats de recherche: Contribution dans un livre/un catalogue/un rapport/dans les actes d'une conférence › Article dans les actes d'une conférence/un colloque

TY - GEN

T1 - Geometry of goodness-of-fit testing in high dimensional low sample size modelling

AU - Marriott, Paul

AU - Sabolova, Radka

AU - van Bever, Germain

AU - Critchley, Frank

PY - 2015

Y1 - 2015

N2 - We introduce a new approach to goodness-of-fit testing in the high dimensional, sparse extended multinomial context. The paper takes a computational information geometric approach, extending classical higher order asymptotic theory. We show why the Wald – equivalently, the Pearson χ2 and score statistics – are unworkable in this context, but that the deviance has a simple, accurate and tractable sampling distribution even for moderate sample sizes. Issues of uniformity of asymptotic approximations across model space are discussed. A variety of important applications and extensions are noted.

AB - We introduce a new approach to goodness-of-fit testing in the high dimensional, sparse extended multinomial context. The paper takes a computational information geometric approach, extending classical higher order asymptotic theory. We show why the Wald – equivalently, the Pearson χ2 and score statistics – are unworkable in this context, but that the deviance has a simple, accurate and tractable sampling distribution even for moderate sample sizes. Issues of uniformity of asymptotic approximations across model space are discussed. A variety of important applications and extensions are noted.

UR - http://www.scopus.com/inward/record.url?scp=84950336625&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-25040-3_61

DO - 10.1007/978-3-319-25040-3_61

M3 - Conference contribution

AN - SCOPUS:84950336625

SN - 9783319250397

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 569

EP - 576

BT - Geometric Science of Information - 2nd International Conference, GSI 2015, Proceedings

A2 - Nielsen, Frank

A2 - Barbaresco, Frederic

A2 - Nielsen, Frank

PB - Springer Verlag

T2 - 2nd International Conference on Geometric Science of Information, GSI 2015

Y2 - 28 October 2015 through 30 October 2015

ER -

Marriott P, Sabolova R, van Bever G, Critchley F. Geometry of goodness-of-fit testing in high dimensional low sample size modelling. Dans Nielsen F, Nielsen F, Nielsen F, Barbaresco F, Barbaresco F, Nielsen F, rédacteurs en chef, Geometric Science of Information - 2nd International Conference, GSI 2015, Proceedings. Springer Verlag. 2015. p. 569-576. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-25040-3_61

Geometry of goodness-of-fit testing in high dimensional low sample size modelling

Résumé

Série de publications

Une conférence

Accès au document

Autres fichiers et liens

Empreinte digitale

Contient cette citation